def test_contains_scalar():
t = Table()
t.add_column('a', np.arange(4))
t.add_column('b', np.arange(2))
filt = t.a.contains(np.array([11]))
assert np.all(filt.values == np.array([False, False, False, False]))
assert np.all(filt.index == np.array([1, 1, 1, 1]))
def test_add_one():
tb = Table({
'a': pd.date_range('2000-01-01', freq='M', periods=10),
'b': np.random.randn(10)
})
tb.add_column('schedule', np.array(['first']))
assert np.all(tb.index == np.array([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 0, 0, 0, 0, 0, 0, 0, 0, 0]]))
def test_add_align():
t = Table()
t.add_column('a', [1, 2, 3, 4, 5, 6], dtype=np.int64)
t.add_column('b', [1, 2, 3], align='bottom', dtype=np.int64)
c = t.a + t.b
assert np.all(c.values == np.array([5, 7, 9]))
assert t.a.values.dtype == np.int64
assert c.values.dtype == np.int64
def test_filter_3():
t = Table()
t.a = np.random.rand(10)
t.b = pd.date_range('2000-01-01', freq='M', periods=10)
t.c = np.array([1, 2])
t.add_column('d', np.array([1, 2]), align='bottom')
t1 = t.filter(t.c > 0)
assert t1.c.values[0] == 1
assert np.all(t1.d.values == np.array([]))
def test_hcat_table():
t = Table({'a': [1, 2, 3], 'b': np.array([4, 5, 6])})
t.add_column('c', [7, 8, 9])
assert np.all(t.index == np.ones((3, 3), dtype=np.uint8))
assert np.all(t.c.values == np.array([7, 8, 9]))
t.add_column('d', [0, 1, 2, 3, 4])
assert np.all(t.index == np.array(
[[1, 1, 1, 0, 0], [1, 1, 1, 0, 0], [1, 1, 1, 0, 0], [1, 1, 1, 1, 1]],
dtype=np.uint8))
def test_sort_0():
t1 = Table()
t1.add_column('a', [1, 2, 3, 4, 5])
t1.add_column('b', [1, 2, 3])
t2 = t1.copy()
t3 = t1.copy()
t4 = Table.from_chunks([t1, t2, t3])
assert np.all(
t4.b.index == np.array([1, 1, 1, 0, 0, 1, 1, 1, 0, 0, 1, 1, 1, 0, 0]))
def test_crop_2():
t = Table()
t.a = np.random.rand(10)
t.b = pd.date_range('2000-01-01', freq='D', periods=10)
t.c = np.array([1, 2])
t.add_column('d', np.array([1, 2]), align='bottom')
t1 = t.crop('d')
assert np.all(t1.c.values == np.array([]))
assert np.all(t1.d.values == np.array([1, 2]))
assert np.all(t1.a.values == t.a.values[-2:])
assert np.all(t1.b.values == t.b.values[-2:])
def test_vcat_heterogeneous():
tb = Table({
'a': pd.date_range('2000-01-01', freq='M', periods=3),
'b': np.random.randn(3)
})
tb.add_column('schedule', np.array(['first']))
tb1 = tb.copy()
tb1.schedule.values[0] = 'second'
tb.stack(tb1)
assert np.all(tb.index == np.array(
[[1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1], [1, 0, 0, 1, 0, 0]],
dtype=np.uint8))
assert np.all(tb.schedule.values == np.array(['first', 'secon']))
def test_merge_empty_1():
table_a = Table()
table_b = Table()
table_a.add_column('a', [])
table_a.add_column('b', [])
table_b.add_column('a', [1, 1, 1])
table_b.add_column('c', [1, 2, 3])
c = merge(table_a, table_b, 'a')
assert np.all(c.a.index == np.array([1, 1, 1]))
assert np.all(c.b.index == np.array([0, 0, 0]))
assert np.all(c.c.index == np.array([1, 1, 1]))
def test_outer_join_datetime():
t1 = Table({
'a': pd.date_range('2000-01-01', periods=20),
'b': np.arange(20)
})
t2 = Table({
'a': pd.date_range('2000-01-01', periods=10),
'b': np.arange(10)
})
t2.add_column('c', [])
t3 = full_outer_join(t1, t2, 'a')
assert np.all(t3.b.values == np.arange(20))
def test_filter_contains_1():
t = Table()
t.add_column('a', np.arange(10))
t.add_column('b', np.arange(5))
filt = t.a.contains(t.b)
assert np.all(t.filter(filt).a.values == t.b.values)
filt = t.a.contains(t.b.values)
assert np.all(t.filter(filt).a.values == t.b.values)
assert len(t.filter(filt)) == 5
filt = t.a.contains(np.array([11]))
assert t.filter(filt).a.is_empty()
def test_inner_join_1():
t1 = Table()
t1.add_column('a', [1, 2, 3, 4, 5, 6])
t1.add_column('b', [1, 2, 3])
t2 = Table()
t2.add_column('a', [2, 3, 4])
t2.add_column('b', [1, 1, 1])
t2.add_column('c', [5, 6, 7])
t3 = inner_join(t1, t2, 'a')
assert np.all(t3.a.values == np.array([2, 3, 4]))
assert np.all(t3.b.values == np.array([2, 3]))
assert np.all(t3.c.values == np.array([5, 6, 7]))
def test_filter_date():
t = Table()
t.a = np.random.rand(10)
t.b = pd.date_range('2000-01-01', freq='D', periods=10)
t.c = np.array([1, 2])
t.add_column('d', np.array([1, 2]), align='bottom')
thres1 = np.array(['2000-01-03'], dtype=np.datetime64)
thres2 = np.array(['2000-01-05'], dtype=np.datetime64)
t1 = t.filter(t.b >= thres1)
assert np.all(t1.c.values == np.array([]))
assert np.all(t1.d.values == np.array([1, 2]))
assert np.all(t1.a.values == t.a.values[2:])
t1 = t.filter((t.b >= thres1) & (t.b <= thres2))
assert np.all(t1.c.values == np.array([]))
assert np.all(t1.d.values == np.array([]))
assert np.all(t1.a.values == t.a.values[2:5])
t1 = t.filter(t.b.date_range(fr=thres1, to=thres2))
assert np.all(t1.c.values == np.array([]))
assert np.all(t1.d.values == np.array([]))
assert np.all(t1.a.values == t.a.values[2:5])